Nextflow workflow for taxonomic classification of shotgun metagenomic data with Kraken 2.
This workflow performs taxonomic profiling through lowest common ancestor (LCA) positioning approach using Kraken2. In this approach the pre-aligned sequences are hierarchically classified on a taxonomy tree using a placement algorithm. It requires database for comparison with the appropriate taxonomical information so this workflow comes packaged together with the MiniKraken Database.
This workflow is implemented in 3 data processing steps: QC, Taxonomic Profiling and Visualization. For each step the following tools are used:
Release the Kraken workflow expects a single sample file (FASTQ), passed with the --fastq
parameter, and generates an HTML Krona plot with the taxonomic information. The Kraken report is stored under results/kraken
.
nextflow main.nf --fastq="test_data/SRR8886136_{1,2}*"